Default layout has chords overlapping voltas

• May 27, 2014 - 08:31
Type
Functional
Severity
S4 - Minor
Status
closed
Regression
No
Workaround
No
Project

To reproduce:
add a volta over a bar
add some chord symbols over the same bar

:( hopefully this doesn't require something of the magnitude of Sibelius' magnetic layout to work around.


Comments

Realistically, to solve this problem in general, yes it does require something like "magnetic layout" - a very complex and expensive operation unlikely to be implemented any time soon.

However, if you know a score is going to have chord symbols, there is a pretty easy solution, - simply define the default height of chord symbols and voltas in Style / General to clear each other. At most, they might still overlap a little bit horizontally, but then only a slight nudge with the keyboard (click chord, ctrl-right) would allow them to clear.

Most likely, the templates for lead sheets and other types of scores that typically include chord symbols will do this by default.

Similar story with tempo except here, you can actually define the default horizontal position for tempo markings in the text style as well. It has been suggested tempo markings at the beginning of a score should automatically align further to left (eg, over the time signature rather than the first note); this will also help if implemented.

Is it worth the effort of coming up with a reference score that MuseScore's defaults should aim to have no conflicting elements in? To me, I cannot see a reason why the two changes you suggest should not be defaults. I am happy to work towards this if it's worthwhile.

I like the idea of a reference score, and was planning on using something like that when I look at #25247: Clean up default style settings, the task I created a while back to make sure we remember to make a pass through the default style settings before release.

As for the specific settings you mention, though, they are not likely to change for 2.0, for good and not so good reasons.

The good reason has to do with voltas. You have to realize that scores with chord symbols are in the minority. Most scores don't contain them, and for these scores, placing voltas high enough to clear non-existent chord symbols makes no sense. It would just waste space. Only scores with chord symbols need voltas that high. So that's why I said I plan to do that in the relevant templates only.

For tempo markings, I think we all agree they belong over the time signature, not over the note the are attached to. But MuseScore has no concept of attaching markings to time signatures. Text is applied to notes. Someday we may extend MuseScore to allow text to be applied to time signatures or other elements, or provide an additional style option to force them to align with the preceding time signature even when applied to a note. But either way, it can't be done just by tweaking the default style. It will require semi-significant code changes and it's too close to release for this to be likely. I will say I haven't given up on the idea, but I think the prevailing opinion is that this is not important enough to delay release of 2.0 ocer, and I can accept that. So I'm prepared to wait until 2.1 or whatever if that feels better. Meanwhile, it takes only a couple of clicks to nudge the tempo over.

Thanks for the in-depth response. :) What about just having chord symbols above the voltas in the default score? I have seen quite a few charts written this way, including older hand-engraved ones. It's non-intrusive to charts without chord symbols, it's better-than-collision for the case of adding chord symbols to a non-template score, and as you say, the defaults for e.g. a lead sheet template could still be mucked with to have symbols underneath.
W.R.T. tempos, would an ugly hack for "if at the start of the score, add a slight left offset at time of creation" be acceptable in the short term? Doesn't solve the general case but would handle (I would hazard) a majority of the commonly seen scenarios (tempo at the start should be offset while later should be more-or-less above the relevant point of change). Surely it's still an improvement on the current placement, and I can't think of much in the way of side effects.

These are definitely ideas I've considered. I am not convinced either is good though.

It is true that setting defaults so that chord symbols are above voltas solves one problem (avoiding collisions) but now it creates another: chord symbols are unnecessarily high. This is no better than having voltas unnecessarily high. Plus you end up with a decidedly non-standard score. While there may exist examples of scores that place chord symbols above voltas, it isn't the norm. And if you want to fix your score to look standard, you'd have to change *both* the volta *and* the chord symbol defaults. So unfortunately, I just don't see a way to have a single one-size-all set of style settings to solve this problem - different scores *will* need different style settings, that's just a fact of life. That's why I keep pushing templates as the best solution we can offer. The different templates can have appropriate style settings *for that type of score*. Also, you can in 2.0 save any group of style settings you want as your own personal default. So if you decide you are happy with chord symbols above voltas by default, you can make it so - without imposing that on anyone else.

It's a different story with tempo. There is no doubt that it is more correct to align tempo with time signatures than with the first note of a measure, if the measure has a time signature. This is not something that is different for different types of scores, so it *would* be possible to find a one-size-fits-all solution. And the possibility you suggest is an obvious candidate for accomplishing it without a ton of trouble. But how much to move the tempo to the left? That will depend on a whole lot of factors. And realistically, unless your tempo marking is short, there still a good change it's going to collide with a chord symbol on the first note. So you're going to need user intervention as often as not anyhow. So the question becomes, is it even worth the trouble?

Really, though, independently of how possible it is to implement, it is still an "ugly hack", and I think it also comes down to whether or not we want to *support* that ugly hack forever. I think the prevailing opinion is, better to solve the problem *well* sometime later than solve it *poorly* for now and then have to support that poor solution indefinitely. As I said, it's trivially simple to move the tempo marking yourself, and since we are basically just talking about tempo changes at start of score (or at other places where there are time signatures or perhaps other elements at the beginning of the measure), it's not like you are going to lose a lot of time making the adjustment. But again, I haven't totally given up on this for 2.0 - I'm just trying to be pragmatic.